The identification of genetic markers for non-insulin dependent diabetes (NIDDM) would permit a full evaluation of the role of non-genetic factors in the expression of that common disease, and thus the possible prevention of inherited diabetes and its complications. This proposal is to examine the genetic predisposition to NIDDM by application of the tools of molecular genetics to Caucasian pedigrees ascertained through an NIDDM or gestational diabetic proband with one affected sibling. Pedigree members will be screened for diabetes and glucose intolerance by standard 75 gm oral glucose tolerance test. Inherited patterns of insulin levels and insulin secretion will be examined to identify phenotypic markers of genetic heterogeneity, and lymphoblastoid cell lines will be established and DNA prepared. These pedigrees will be examined for defects of the insulin receptor gene, the insulin gene, and the glucose transporter gene by using previously identified DNA polymorphisms or polymorphisms to be identified for each locus to test for linkage to diabetes. Where linkage is identified, haplotypes defined by these polymorphisms will be used to examine the role of environmental factors and obesity in predisposed individuals not expressing NIDDM. The putatively defective genes will be characterized at the molecular level by cloning and sequencing. This analysis will provide an estimate of the role of these genes in greater than 5% of pedigrees with inherited diabetes. The clinical data will be subjected to segregation analysis to identify subgroups with uniform mode of inheritance and phenotypic markers, which would be suitable for a general mapping approach using random polymorphic DNA markers if no major predisposing gene is discovered with the candidate approach.